Problems with Rosetta version 5.41

Message boards : Number crunching : Problems with Rosetta version 5.41

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · Next

AuthorMessage
FluffyChicken
Avatar

Send message
Joined: 1 Nov 05
Posts: 1260
Credit: 369,635
RAC: 0
Message 32009 - Posted: 3 Dec 2006, 14:47:17 UTC
Last modified: 3 Dec 2006, 15:12:44 UTC

Chu, maybe it is time to send the debugging files down like we used to in the early months of Rosetta@home.

I would recommend doing that over at Ralph (unless it still is) and getting people to test over there.
Also to do it over here it was a seperate <filename>.pdb


Then and a definate must is to report this problem to the project development list.

I know we think it is graphics card based, you could actually start collecting that information if you updated your servers. Ok so only peopl using the ner >5.5.x, 5.6.x and 5.7.x (soon to be 5.8.x ) will give you the graphics card type, but it is feedback you could use.

I would suggest all of the people having trouble to test out the 5.7.5 client, linux users should also have that version soon as well.
I know if the problems are still happening with that the developers would like the feedback.

Also everyone having problems shoudl try 'resetting' Rosetta@home (on projects tab, select rosetta and hit the reset button. It clears everythign out and get new version of all the files.

Honestly though screensaver problems have been going on for a while and people are leaving because of it. So 'Project Devs' go ask the 'Boinc devs' on the list they provide, other project may have found a solution to it.

BOINC developers list http://www.ssl.berkeley.edu/mailman/listinfo/boinc_dev]
BOINC Projects list http://www.ssl.berkeley.edu/mailman/listinfo/boinc_projects

All BOINC versions http://boinc.berkeley.edu/download_all.php


It would be nice to have these problems sorted before BOINC devs get 5.8 out of the door, which will hopfully be befroe christmas, ready to pick up the super powerful computer people send silly money on to use MS Word :D, we may as well use some of that power.




Team mauisun.org
ID: 32009 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Feet1st
Avatar

Send message
Joined: 30 Dec 05
Posts: 1755
Credit: 4,690,520
RAC: 0
Message 32013 - Posted: 3 Dec 2006, 17:50:56 UTC - in response to Message 32005.  

can anyone tell me why i got all these errors???????????????/


For anyone to try and give you an answer to that, you will have to report what errors you are talking about. You can select them from the messages tab in the BOINC Manager, then copy them and paste them into this forum.
Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/
ID: 32013 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rochester new york
Avatar

Send message
Joined: 2 Jul 06
Posts: 2842
Credit: 2,020,043
RAC: 0
Message 32014 - Posted: 3 Dec 2006, 17:55:27 UTC - in response to Message 32005.  

can anyone tell me why i got all these errors???????????????/

https://boinc.bakerlab.org/rosetta/result.php?resultid=49351166
ID: 32014 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Feet1st
Avatar

Send message
Joined: 30 Dec 05
Posts: 1755
Credit: 4,690,520
RAC: 0
Message 32015 - Posted: 3 Dec 2006, 18:02:33 UTC

Chu

It seems to me as though there was a fairly noticeble increase in reported problems with graphics at about the same time as the project released the code to show the sidechains. Do you have statistics to support that assertion? Does the sidechain code use any different graphics routines then the original graphics did? Would there be any way to run some tests on Ralph with and without sidechains displayed and see if there is indeed a correlation? The preferences allow you to set parameters for how many frames per second and maximum CPU for graphics etc. I've seen no discussion of how people have these parameters set. But it would seem relevant.

I agree with fluffy, the way to lick this is with study on Ralph. And since many Ralph users already have disabled the graphics, or run as a service, you'd have to specially request that people enable it in the way that you wish to study (and give them some time to see your message, and get to their machines to set things as desired for the test). And send WUs on Ralph until we gather enough numbers on what works and what doesn't until we nail this down.

There's been mention of ATI cards, but do we have any numbers on ATI as compared to other? Maybe 99% of the ATI folks are quietly running fine.
Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/
ID: 32015 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Feet1st
Avatar

Send message
Joined: 30 Dec 05
Posts: 1755
Credit: 4,690,520
RAC: 0
Message 32016 - Posted: 3 Dec 2006, 18:10:16 UTC - in response to Message 32014.  

can anyone tell me why i got all these errors???????????????/

https://boinc.bakerlab.org/rosetta/result.php?resultid=49351166


This work unit was ended by the watchdog. This is a sort of a failsafe mechanism that Rosetta has implemented to assure that your machine doesn't get caught in a loop in the twists and turns of the model you are working on.

The WU reported with validation errors and was for Rosetta v5.40. As discussed here in the details of the v5.41 Rosetta release, this problem has now been resolved. Your machine will be getting WUs for v5.41 now as it requests work, so there's really nothing you need to do on your end. The thread you are posting in here is for issues with the v5.41 version of Rosetta. You can see the version in the BOINC manager tasks tab, or, it is shown on the website at the very bottom for completed WUs. If you see a similar problem with v5.41, please report it here in this thread we're in right now.
Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/
ID: 32016 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile hedera
Avatar

Send message
Joined: 15 Jul 06
Posts: 76
Credit: 5,139,863
RAC: 905
Message 32022 - Posted: 3 Dec 2006, 19:51:42 UTC

FluffyChicken's mention of ATI cards made me go check - not being a gamer I don't especially care what my video card is - and yes, I have an ATI Radeon X300SE video card. I've had no crashes since I turned off the BOINC screen saver. I'd be happy to participate in testing if someone would tell me how to set it up. I've just reset the Rosetta project as instructed; I'll turn the screen saver back on and report results. I'm still seeing 5.41.

Let me know if you need details of the video driver I'm using - I don't believe I've ever updated the video drivers.

Also, per Feet1st's remark about frames per second and graphics preferences, I am using the system default preferences entirely.
--hedera

Never be afraid to try something new. Remember that amateurs built the ark. Professionals built the Titanic.

ID: 32022 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rochester new york
Avatar

Send message
Joined: 2 Jul 06
Posts: 2842
Credit: 2,020,043
RAC: 0
Message 32029 - Posted: 3 Dec 2006, 20:49:05 UTC - in response to Message 32016.  
Last modified: 3 Dec 2006, 20:49:35 UTC

can anyone tell me why i got all these errors???????????????/

https://boinc.bakerlab.org/rosetta/result.php?resultid=49351166


This work unit was ended by the watchdog. This is a sort of a failsafe mechanism that Rosetta has implemented to assure that your machine doesn't get caught in a loop in the twists and turns of the model you are working on.

The WU reported with validation errors and was for Rosetta v5.40. As discussed here in the details of the v5.41 Rosetta release, this problem has now been resolved. Your machine will be getting WUs for v5.41 now as it requests work, so there's really nothing you need to do on your end. The thread you are posting in here is for issues with the v5.41 version of Rosetta. You can see the version in the BOINC manager tasks tab, or, it is shown on the website at the very bottom for completed WUs. If you see a similar problem with v5.41, please report it here in this thread we're in right now.

////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////ok thanks i just want to make sure im doing any good here
ID: 32029 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile hedera
Avatar

Send message
Joined: 15 Jul 06
Posts: 76
Credit: 5,139,863
RAC: 905
Message 32033 - Posted: 3 Dec 2006, 22:16:29 UTC
Last modified: 3 Dec 2006, 22:17:27 UTC

Reporting results of latest experiment:

Having reset Rosetta and restarted everything with fresh downloads at 11:49:12, I turned the BOINC screen saver back on, still running production 5.41. The next work unit started:

12/03/2006 11:49:55 AM|rosetta@home|Starting task s018__BOINC_ABRELAX_SAVE_ALL_OUT_hom002__1407_17911_0 using rosetta version 541

At 2:07:37 I got a hang on that work unit:


12/03/2006 2:07:37 PM|rosetta@home|Unrecoverable error for result s018__BOINC_ABRELAX_SAVE_ALL_OUT_hom002__1407_17911_0 ( - exit code 1073807364 (0x40010004))


Here is the Windows event log:

Hanging application rosetta_5.41_windows_intelx86.exe, version 0.0.0.0, hang module hungapp, version 0.0.0.0, hang address 0x00000000.

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

Fault bucket 353414255.

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

I believe this is the same hang I've been getting. I'm back to the other screen saver now. Again, let me know how and I'll be happy to help test.
--hedera

Never be afraid to try something new. Remember that amateurs built the ark. Professionals built the Titanic.

ID: 32033 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
P . P . L .

Send message
Joined: 20 Aug 06
Posts: 581
Credit: 4,865,274
RAC: 0
Message 32040 - Posted: 4 Dec 2006, 1:40:56 UTC

Don't know if this is a problem, just returned my first 5.41 and there

is a lot in the file is it normal.

https://boinc.bakerlab.org/rosetta/result.php?resultid=49984017

ID: 32040 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
EW-3

Send message
Joined: 1 Sep 06
Posts: 27
Credit: 2,561,427
RAC: 0
Message 32052 - Posted: 4 Dec 2006, 3:52:08 UTC - in response to Message 32022.  

I've had several problems whenever I let the screen saver run, and it goes back to prior versions.
But the common item is I have a ATI Radeon Express 200 series interface.
just more data for the analysis.


FluffyChicken's mention of ATI cards made me go check - not being a gamer I don't especially care what my video card is - and yes, I have an ATI Radeon X300SE video card. I've had no crashes since I turned off the BOINC screen saver. I'd be happy to participate in testing if someone would tell me how to set it up. I've just reset the Rosetta project as instructed; I'll turn the screen saver back on and report results. I'm still seeing 5.41.

Let me know if you need details of the video driver I'm using - I don't believe I've ever updated the video drivers.

Also, per Feet1st's remark about frames per second and graphics preferences, I am using the system default preferences entirely.


ID: 32052 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Seventh Serenity

Send message
Joined: 30 Nov 05
Posts: 18
Credit: 87,811
RAC: 0
Message 32073 - Posted: 4 Dec 2006, 20:54:23 UTC

It appears there is something wrong with the Linux client on my Kubuntu Dapper Linux machine as the new Rosetta app keeps crashing, but LHC@Home and Einstein@Home are working flawlessly. The machine is using kernel 2.6.15 and is optimised for i686 processors. Here is the main spec of the machine:

* Gigabyte 8S648FXP-RZ SiS 648FX Motherboard
* Pentium 4 Northwood Extreme Edition @ 3.2GHZ
* 1.5GB PC3200 DDR Kingston Value RAM (3x 512MB, Single Channel)

I really wish this system could run on Rosetta@Home - that P4 is a monster CPU.

Any ideas?
"In the beginning the universe was created. This made a lot of people very angry and is widely considered as a bad move." - The Hitchhiker's Guide to the Galaxy
ID: 32073 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Chu

Send message
Joined: 23 Feb 06
Posts: 120
Credit: 112,439
RAC: 0
Message 32076 - Posted: 4 Dec 2006, 21:30:00 UTC - in response to Message 32040.  

It looks fine to me and those extra lines are probably from testing new energy tables.
Don't know if this is a problem, just returned my first 5.41 and there

is a lot in the file is it normal.

https://boinc.bakerlab.org/rosetta/result.php?resultid=49984017


ID: 32076 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Chu

Send message
Joined: 23 Feb 06
Posts: 120
Credit: 112,439
RAC: 0
Message 32077 - Posted: 4 Dec 2006, 21:34:14 UTC - in response to Message 32073.  

Thanks for posting this, Electrolyte. Hopefully with the help from all the experts here, the problem can be solved. Sorry for any inconvenience the new application has brought to you.
It appears there is something wrong with the Linux client on my Kubuntu Dapper Linux machine as the new Rosetta app keeps crashing, but LHC@Home and Einstein@Home are working flawlessly. The machine is using kernel 2.6.15 and is optimised for i686 processors. Here is the main spec of the machine:

* Gigabyte 8S648FXP-RZ SiS 648FX Motherboard
* Pentium 4 Northwood Extreme Edition @ 3.2GHZ
* 1.5GB PC3200 DDR Kingston Value RAM (3x 512MB, Single Channel)

I really wish this system could run on Rosetta@Home - that P4 is a monster CPU.

Any ideas?


ID: 32077 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Chu

Send message
Joined: 23 Feb 06
Posts: 120
Credit: 112,439
RAC: 0
Message 32078 - Posted: 4 Dec 2006, 21:41:00 UTC - in response to Message 32009.  

Ooops, thanks for pointing this out, FluffyChicken. I thought we had added the pdb symobol for degbugging already and it turns out the files were loaded into the correct place, but with inproper file permission. The problem has been fixed now. With the help from Rom Walton, it has been a routine for the project team to add the pdb debugging symbol file for every windows build, both on Ralph and here.
Chu, maybe it is time to send the debugging files down like we used to in the early months of Rosetta@home.



ID: 32078 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Feet1st
Avatar

Send message
Joined: 30 Dec 05
Posts: 1755
Credit: 4,690,520
RAC: 0
Message 32079 - Posted: 4 Dec 2006, 22:05:07 UTC

Chu, could you describe how to get and use the PDB? And any other ideas or observations you have on the screensaver issues? Do you have any data to support or refute my ideas from last week?

I've created a new thread for that topic, because it does not appear directly release dependant.

I'm hoping you might add in to that thread some steps and ideas about how to CAUSE the screen saver problems to occur. Is it just "screensaver"? or is it any time Rosetta graphics are displayed? If we can catch some with the PDB, perhaps enough detail can be gathered to figure out the problem.
Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/
ID: 32079 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Chu

Send message
Joined: 23 Feb 06
Posts: 120
Credit: 112,439
RAC: 0
Message 32082 - Posted: 4 Dec 2006, 22:19:24 UTC - in response to Message 32015.  
Last modified: 4 Dec 2006, 22:20:01 UTC

Thanks for the suggestions, FluffyChicken and Feet1st.

I agree that it seems that more graphic-related problems have been reported since the Rosetta started to show sidechains. In the code, sidechains drawing happens at the same place as the original backbone trace drawing, however, as obviously seen, it requires more resource from the graphic thread to do that because for each residue the number of atoms to be drawn increases from 1 (C-alpha) to on average of 10 or more. Back to the time when tested sidechain drawing on Ralph, we did find some problems in which the graphic routine did not like share variables and numbers too much with protein simulation routines. After minimizing the usage of sharing, the error rate did drop a lot, but I guess the so-far increasing graphic-related error rate is correlated with the inevitablly increasing number of atoms we are showing on the screen. The project team will discuss about this issue and lay out a plan on how to solve this problem. We will definitely try to seek help from the boinc developer list and experts on graphic software engineering. Meanwhile, we will keep collecting info from clients regarding this issue to see if there is any clue.

Thanks for everyone who has reported the graphic-related problems and are willing to help us to test to have this problem solved.
Chu

It seems to me as though there was a fairly noticeble increase in reported problems with graphics at about the same time as the project released the code to show the sidechains. Do you have statistics to support that assertion? Does the sidechain code use any different graphics routines then the original graphics did? Would there be any way to run some tests on Ralph with and without sidechains displayed and see if there is indeed a correlation? The preferences allow you to set parameters for how many frames per second and maximum CPU for graphics etc. I've seen no discussion of how people have these parameters set. But it would seem relevant.

I agree with fluffy, the way to lick this is with study on Ralph. And since many Ralph users already have disabled the graphics, or run as a service, you'd have to specially request that people enable it in the way that you wish to study (and give them some time to see your message, and get to their machines to set things as desired for the test). And send WUs on Ralph until we gather enough numbers on what works and what doesn't until we nail this down.

There's been mention of ATI cards, but do we have any numbers on ATI as compared to other? Maybe 99% of the ATI folks are quietly running fine.


ID: 32082 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Stevea

Send message
Joined: 19 Dec 05
Posts: 50
Credit: 738,655
RAC: 0
Message 32121 - Posted: 5 Dec 2006, 19:14:34 UTC

1st time I have ever seen this error?

https://boinc.bakerlab.org/rosetta/result.php?resultid=50153650

Is this an error because someone turned in a result before I did? If so there is no reason for me to crunch 8hr WU's, I might as well set all my machines to 1 hr.


CPU time 28876.390625
stderr out


<core_client_version>5.3.12.tx36</core_client_version>
<stderr_txt>
# random seed: 3050933
# cpu_run_time_pref: 28800
======================================================
DONE :: 1 starting structures built 30 (nstruct) times
This process generated 76 decoys from 76 attempts
======================================================


BOINC :: Watchdog shutting down...
BOINC :: BOINC support services shutting down...

</stderr_txt>

Validate state Workunit error - check skipped
Claimed credit 145.341688436918
Granted credit 0
application version 5.41
ID: 32121 · Rating: 1 · rate: Rate + / Rate - Report as offensive    Reply Quote
genes
Avatar

Send message
Joined: 8 Oct 05
Posts: 60
Credit: 496,133
RAC: 1,175
Message 32139 - Posted: 6 Dec 2006, 2:51:14 UTC

Had this one crash while running graphics: resultid=50653149

the usual 0xC0000005 error. I thought I had maybe gotten the problem under control last night, I installed an ATI graphics card instead of the NVidia one that I had been running with for the longest time. It did run one WU with screensaver graphics enabled, but then this morning I found this one frozen. With the NVidia card, the graphics don't last more than a few minutes before freezing.

System: Balrog

Current VGA card: ATI Radeon X800 XL AGP, driver 6-11_xp-2k_dd_37616
Previous VGA card: NVidia GeForce FX5950 AGP, driver 91.31.

I have tried many other drivers for this and other NVidia cards, Rosetta/Ralph has problems with the graphics on all of them so far. I do have another ATI card I can try, I think it's an X850XT. I could also try their CCC driver, though it's bloatware and I don't like it.

If anybody has any suggestions for good NVidia or ATI driver versions, I'm willing to try them out.

ID: 32139 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Tony DeBari

Send message
Joined: 13 Apr 06
Posts: 12
Credit: 2,944,090
RAC: 0
Message 32150 - Posted: 6 Dec 2006, 8:22:13 UTC

Hi all. One more data point re graphics-related failures.

ResultID 50704727

Dual AMD AthlonMP 2000+, 512 MB RAM, ATI Radeon 9250, WinXP Pro (SP2), no screen saver (blank screen)

stderr out looks like this:
<core_client_version>5.4.9</core_client_version>
<message>
- exit code 1073807364 (0x40010004)
</message>
<stderr_txt>
# random seed: 2534060
# cpu_run_time_pref: 21600

</stderr_txt>

I had popped up the graphics display and was rotating and zooming the native structure when the window stopped responding. However, BoincView still showed CPU time accumulating, so I believe the science app was still running. I forced the graphics window to close and it crashed the science app. This is the first time anything like this has happened on this computer.

Regards,

-- Tony
ID: 32150 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
FluffyChicken
Avatar

Send message
Joined: 1 Nov 05
Posts: 1260
Credit: 369,635
RAC: 0
Message 32160 - Posted: 6 Dec 2006, 9:52:32 UTC - in response to Message 32139.  

For the ATI cards could you try 6.5 official.
The type of control centre/panel shouldn't matter since it is unrelated to the driver.


For NVIDIA I have 97.02 running nicely on my GeForce-Go 5200FX, I'll need to check what I have on the other GeForceFX

What motherboard are you using ?


Also make sure you install the latest DirextX http://www.microsoft.com/dirextx Just run it after you install the drivers, it only updates what it needs to update. It is updated noramlly every 2 months so the last update was October.




Had this one crash while running graphics: resultid=50653149

the usual 0xC0000005 error. I thought I had maybe gotten the problem under control last night, I installed an ATI graphics card instead of the NVidia one that I had been running with for the longest time. It did run one WU with screensaver graphics enabled, but then this morning I found this one frozen. With the NVidia card, the graphics don't last more than a few minutes before freezing.

System: Balrog

Current VGA card: ATI Radeon X800 XL AGP, driver 6-11_xp-2k_dd_37616
Previous VGA card: NVidia GeForce FX5950 AGP, driver 91.31.

I have tried many other drivers for this and other NVidia cards, Rosetta/Ralph has problems with the graphics on all of them so far. I do have another ATI card I can try, I think it's an X850XT. I could also try their CCC driver, though it's bloatware and I don't like it.

If anybody has any suggestions for good NVidia or ATI driver versions, I'm willing to try them out.


Team mauisun.org
ID: 32160 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Previous · 1 · 2 · 3 · 4 · 5 · Next

Message boards : Number crunching : Problems with Rosetta version 5.41



©2024 University of Washington
https://www.bakerlab.org